And what you do suggest as an alternative? I actually hesitate to use anything too avant-garde since I'm a pretty risk-averse sort of person. Plus, at the end of the day it is just a guess... basically. Ok, maybe a well-informed guess, but still. Not to mention, the data providers are doing most of the heavy lifting... I'm pretty finicky with data though. I certainly do have a db for myself, but I compare providers, I shop around. It's fun, lol. EWMA was by far easier to implement but garch isn't exactly that bad... there are good libraries out there. The hardest part was wiring the data in and testing/QA, but I'm pretty good at that kind of thing since I've been doing it since my mid-20s.
Yes, true, but you just have to accept that Bid and Ask (of the option) are the drivers. You can go in and put a better offer, and by this action control the values (ie. a new Bid or Ask and this then changing the IV)...
You mean quote stuffing... not a great long-term practice. I think you'll eventually get nailed for doing that kind of thing. Hard to pull off at a retail level too. Not to mention, if you're not extremely careful, you might cross with someone who knows more than you. Better to just do an honest day's job IMHO.
It depends on your use case. As you said "they're too slow to actually implement in a trading system", one needs fast algorithms when scanning hundreds/thousands of strikes... I would suggest to use the MidPrice. But you have to take care for the value zero. Ie. if both Bid and Ask are zero then there is no way to be able to compute the IV, otherwise if only one of them is zero then of course don't divide by 2. The resulting price you have to feed to the IV finder algorithm. I don't know your programming language, but you can find source code on the net for many languages. If not then simply port it from C to your language as C is IMO simple to understand.
There is nothing unethical in this: you just make your own offer, nothing wrong with that. Normally one does not need to do this step. I mentioned it only to practically show you how IV changes by a new top-of-the-book Bid/Ask.
Lol... starts thread named "illegal option orders in the orderbook", asked about quote stuffing and replies "There is nothing unethical in this".
What a misunderstanding by you. I was meaning to place just one order to the top-of-the-book to change the current best Bid or Ask, and with it changing the IV, whereas quote stuffing is criminal HFT behavior consisting of countless orders, much like a (D)DoS attack.
I use to trade options with spreads $1x$5 and it’s not worth your time. In the past the stock would go in the money with months of time-value. $55 stock price, $50 June call and they would not buy at $5.00. The stock hit $55.30, no buyer at $5. That was years ago, the algos would buy now. I made a video of trying to sell the calls below intrinsic because some traders said that was impossible. Now I avoid them, when Apple trades pennies why bother with wide spreads?
I know earth... but I'm pointing out that there's a contradiction in your thinking overall. If there's nothing wrong with what you suggest, why would there be anything "illegal" about the bid/ask spread? That said, I think it may just be that English isn't your native language. Perhaps the thread should have been titled something like "Why are there peculiar or nonsensical bid/asks in the order book" instead? There's nothing illegal about them.